Use of Microbial Biofilms to Monitor the Efficacy of a Stormwater … · 2012. 7. 26. · Use of...

Use of Microbial Biofilms to Monitor the

Efficacy of a Stormwater Treatment Train September 2009 TR 2009/086

Auckland Regional Council

Technical Report No.086 September 2009

ISSN 1179-0504 (Print)

ISSN 1179-0512 (Online)

ISBN 978-1-877528-98-9

Recommended Citation: Lear, G., Ancion, P., Roberts, K. & G. D. Lewis. (2009). Use of Microbial Biofilms to

Monitor the Efficacy of a Stormwater Treatment Train. Prepared by the University of

Auckland for Auckland Regional Council. Auckland Regional Council Technical Report

2009/086.

© 2008 Auckland Regional Council

This publication is provided strictly subject to Auckland Regional Council's (ARC) copyright and other

intellectual property rights (if any) in the publication. Users of the publication may only access, reproduce and

use the publication, in a secure digital medium or hard copy, for responsible genuine non-commercial

purposes relating to personal, public service or educational purposes, provided that the publication is only

ever accurately reproduced and proper attribution of its source, publication date and authorship is attached to

any use or reproduction. This publication must not be used in any way for any commercial purpose without

the prior written consent of ARC. ARC does not give any warranty whatsoever, including without limitation,

as to the availability, accuracy, completeness, currency or reliability of the information or data (including third

party data) made available via the publication and expressly disclaim (to the maximum extent permitted in

law) all liability for any damage or loss resulting from your use of, or reliance on the publication or the

information and data provided via the publication. The publication and information and data contained within

it are provided on an "as is" basis.

Reviewed by: Approved for ARC Publication by:

Name: Dr Martin Neale Name: Grant Barnes

Position: Project Leader - Freshwater Position: Group Manager – Monitoring and

Research

Organisation: Auckland Regional Council Organisation: Auckland Regional Council

Date: 2nd September 2009 Date: 7th September 2009

Use of Microbial Biofilms to Monitor the Efficacy of a Stormwater Treatment Train

Gavin Lear

Pierre-Yves Ancion

Kelly Roberts

Gillian D. Lewis

Prepared for

Auckland Regional Council

June 2009

Stream Biofilm Research Group,

School of Biological Sciences,

The University of Auckland,

Auckland 1010,

New Zealand

www.streambiofilm.org.nz

Contents

1111 Executive SummaryExecutive SummaryExecutive SummaryExecutive Summary 1111

2222 IntroductionIntroductionIntroductionIntroduction 3333

2.1 Background 3

2.2 Stormwater Sources 3

2.3 The Albany Busway: A Case Study in Stormwater Treatment 4

2.3.1 Source Control and use of catchpits 6

2.3.2 Raingardens and Swales 6

2.3.3 Filters 6

2.3.4 Wetland 7

2.3.5 Lucas Creek 7

2.3.6 Bacterial Communities as an Indicator of Water Quality 8

2.3.7 Accumulation of Metals in Microbial Biofilms 9

2.4 Aims and Objectives 9

3333 MethodsMethodsMethodsMethods 11111111

3.1 Site Description 11

3.1.1 Albany Park and Ride Treatment Train 11

3.1.2 Lucas Creek 12

3.2 Sampling Procedure 14

3.2.1 Collection of Physico-Chemical Stream Data 14

3.2.2 Collection and Processing of Biofilm Samples 14

3.3 Processing of Samples for Metal Analysis 15

3.4 Community Fingerprinting of Bacterial Biomass 15

3.5 Statistical Analysis 16

4444 ResultsResultsResultsResults 17171717

4.1 Stormwater Treatment Train 17

4.1.1 Concentrations of Biofilm Associated Metals 17

4.1.2 Bacterial Community Structure 19

4.2 Lucas Creek 20

4.2.1 Physico-Chemical Stream Data 20

4.2.2 Bacterial Community Structure 24

5555 DiscussionDiscussionDiscussionDiscussion 27272727

5.1 Lucas Creek 27

5.1.1 Accumulation of Metals in the Sediment and Biofilm 27

5.1.2 Biofilm Bacterial Community Structure 28

5.1.3 ARISA Methodology 28

5.2 Stormwater Treatment Train 30

5.3 General Conclusions 31

6666 Conclusions and RecommendationsConclusions and RecommendationsConclusions and RecommendationsConclusions and Recommendations 32323232

6.1 Implications and Recommendation for Current Management 32

6.2 Recommendations for Future Research 32

7777 AcknowledgemAcknowledgemAcknowledgemAcknowledgementsentsentsents 34343434

8888 ReferencesReferencesReferencesReferences 35353535

9999 AppendixAppendixAppendixAppendix 38383838

9.1 Extraction of DNA from Biofilm Biomass 38

9.2 Automated ribosomal intergenic spacer analysis of biofilm DNA 38

9.3 Quantitative Methods 39

9.4 Bacterial ARISA Profiles – Jan 2009 – Treatment Train Samples 40

9.5 Bacterial ARISA Profiles – March 2009 – Treatment Train Samples 41

9.6 Bacterial ARISA Profiles – Jan 2009 – Lucas Creek Samples 42

9.7 Bacterial ARISA Profiles – February 2009 – Lucas Creek Samples 43

9.8 Bacterial ARISA Profiles – March 2009 – Lucas Creek Samples 44

9.9 Photograph of treated wood barrier at the western, downstream end of the treatment

wetland 45

9.10 Photograph showing the large mound of disturbed earth south-west of the treatment

wetland 46

Use of Microbial Biofilms to Monitor the Efficacy of a Stormwater Treatment Train 1

1 Executive Summary Stormwater is considered to have the single largest impact on the ecological health of

urban streams within the Auckland Region. The aim of this study was to test the

efficacy of a stormwater ‘treatment train’ in mitigating the environmental impacts of an

open-air car park on the receiving waters of a nearby stream. To achieve this, the

structure of bacterial biofilm communities both upstream and downstream of the site

of stormwater discharge into the receiving stream were documented and used as a

novel indicator of freshwater ecological health. In addition, bacterial communities were

sampled within the stormwater pipes (where traditional biological indicators [i.e., fish

and macroinvertebrates] are not present) to assess the potential ecological impacts of

stormwater at different stages of the treatment train, and to monitor the quality of

stormwater throughout the site.

The site of the Albany Busway Park and Ride was used as a case study for the

appropriate treatment of urban stormwater. A variety of stormwater treatment

strategies have been incorporated throughout the site to provide an integrated

treatment train. These include the installation of grassy swales, raingardens, catchpit

filters, a large StormFilter treatment device and also a treatment wetland. Stormwater

is directed through this treatment train via a network of underground pipes before

discharge into Lucas Creek. Much of this soft-bottomed stream consists of high-value,

low disturbance sites, and the stream receives a high level of community interest.

Lucas Creek contains abundant koura (Paranephrops planifrons) and populations of

banded kokopu (Galaxias fasciatus), long-finned eel (Anguilla dieffenbachii) and red-

finned bully (Gobiomorphus huttoni). However, the surrounding catchment is

undergoing rapid development (creating new residential and commercial zones) in the

nearby areas of Albany Heights, Fairview Heights and Albany Centre. Effective

management strategies are therefore required to minimise the impacts of increasing

urbanisation on the ecological health of Lucas Creek.

We monitored changes in bacterial community structure (a sensitive biological

indicator of ecosystem health) at 10 different locations within Lucas Creek, both

upstream and downstream of the Park and Ride stormwater outlet, and at 9 sites

within the stormwater treatment train. Bacterial community profiles were used to

provide reliable descriptions of community diversity and composition both within and

between all sample sites. A suite a physico-chemical characteristics, including

concentrations of biofilm- and-sediment associated metals (e.g., Cu, Zn, Pb) were also

recorded for each sampling location.

Concentrations of biofilm-associated Pb, Cu and Zn declined throughout the treatment

train. However, concentrations of biofilm-associated Zn remained high (declining to a

minimum of 1.3 g kg-1 biofilm dry wt. at the end of the treatment train, compared to a

maximum of 4.8 g kg-1 biofilm dry wt. detected at one sampling location).

Unexpectedly, concentrations of arsenic, cadmium, chromium and nickel increased in

the later stages of the treatment train (comparing values obtained at the inlet of the

stormfilter, and the outlet from the wetland). Indeed, concentrations of biofilm-

associated nickel reached a maximum of 190 mg kg-1 dry wt. directly downstream of


the StormFilter device. The cause of increased concentrations of these metals closer

to the discharge outlet into Lucas Creek remains unclear.

Significant differences in bacterial community structure were detected between

sections of Lucas Creek located either upstream, or downstream of the stormwater

outlet. However, whilst significant, the extent of these differences was minor. The

bacterial community structure within the upstream sections of Lucas Creek was very

similar to that within the channel of the stormwater outlet. In addition, bacterial

communities within the latter stages of the treatment train were most similar to those

within Lucas Creek, suggesting a modification of the stormwater to provide similar

environmental conditions to within the stream. Concentrations of the metals (Cu, Pb

and Zn) within the sediment of Lucas Creek close to the stormwater outlet remained

within ANZECC (2000) guidelines for the protection of freshwater ecological health and

did not increase in concentration downstream of the stormwater outlet.

Concentrations of Cu, Pb, Zn, As, Cd, Cr and Ni in the stream water of Lucas Creek

also remained within the values for the protection of 95% of species in freshwater

(ANZECC, 2000). Therefore, both the microbial community, and metal data support

that the environmental impacts of the stormwater are reduced throughout the

treatment train, ensuring that the recent development of the Albany Busway park and

ride car park, and adjoining infrastructure, are causing minimal environmental impact

on the receiving waters of Lucas Creek.


2 Introduction

2.1 Background

Stormwater is considered to have the single largest impact on the ecological health of

urban streams in the Auckland region (BCG, 2009). Recognising this, the Auckland

Regional Council is committed to identifying the environmental effects of stormwater

and advocating regional management solutions. Key components of this approach

include;

• Improving the cost-effectiveness of existing stormwater treatment practices

• Evaluating new, innovative approaches for removing chemical contaminants

from stormwater

• Improving understanding of the cause-effect relationship between stormwater

chemical contaminants and effects on life in streams, estuaries and harbours.

In this report, we investigate the efficacy of a treatment train to remove stormwater

contaminants originating from the Albany Park and Ride car park located at the

northern terminus of Auckland’s northern busway. To achieve this, bacterial

communities were used as a novel biological indicator of the ecological impact of

stormwater at different stages of the treatment train. In addition, bacterial community

analysis was used to monitor the effects of the current stormwater discharge on the

receiving waters of Lucas Creek.

2.2 Stormwater Sources

Stormwater is a general term applied to water that has accumulated on land as a result

of precipitation events and is of concern for two main reasons; (i) flood control and

water supply, and (ii) related contaminants carried in the water. The development of

land has increased the area of impermeable surfaces (roads, buildings, etc.) that may

collect pollutants. These then attenuate until entering rivers and streams following

precipitation. The nature of these contaminants is highly variable and site specific. For

example, runoff from roofs may contain elevated concentrations of synthetic organic

compounds and zinc (from galvanised roofs and gutters) while roads and car parks are

major sources of nickel, polycyclic aromatic hydrocarbons (combustions products of

gasoline), zinc (from tyres) and copper (from vehicle brake pads). In addition, fertilisers

used on lawns are a significant source of nitrates and phosphorus, and herbicides may

impact aquatic plant communities in the receiving waters. The complex composition of

urban stormwater means that a multifaceted ‘treatment-train’ approach is frequently

seen as a desirable method to manage the cocktail of contaminants present within

urban stormwater, a method recently advocated by the Auckland Regional Council

(Figure 2.1).


Figure 2.1Figure 2.1Figure 2.1Figure 2.1

Schematic of the stormwater treatment train advocated by the ARC (ARC, 2003).

2.3 The Albany Busway: A Case Study in Stormwater Treatment

The objective of this study was to monitor the efficacy of a treatment train to mitigate

the biological impacts of stormwater on a freshwater stream. The Albany Park and

Ride station (Fig. 2.2) opened in November 2005 as part of the northern busway

transport system running alongside State Highway 1 in the North of Auckland. Albany

station is an ‘offline’ station, meaning that it is not connected to the other stations by a

physically separated bus route (which currently connects stations from Constellation to

Akoranga). It is located within a grassed area in the vicinity of Albany Town Centre, a

young and rapidly expanding commercial area. Albany station has dedicated park and

ride facilities for ~ 600 cars (located at 36o43’18”S, 174o42’45”E), with another 1000 to

be added in later stages to meet future demands.

Different stormwater management strategies are required for the treatment of various

stormwater contaminants. For example, biofiltration methods, such as swales and rain

gardens are highly effective at removing particulate lead, but have little potential to

reduce concentrations of phosphorus and nitrogen within stormwater. For this reason,

a stormwater treatment train has been integrated into the Albany Busway site in an

attempt to mitigate the effects of the Albany Park and Ride car park on the receiving

waters of Lucas Creek. This treatment train, highlighted in Fig. 2.2, includes

raingardens, grassy swales and engineered wetlands, as well as more engineered

solutions such as Enviropod™ catchpit filters and a 148-cartridge StormFilter (installed

by StormWater 360, for more details refer to www.stormwater360.co.nz). Details of

each component of the treatment train are provided below.


Figure 2.2:Figure 2.2:Figure 2.2:Figure 2.2:

Map of the Albany Busway site (36o43’18 S, 174o42’45 E) showing stormwater ‘treatment train’.

Adapted from a map produced by M. Ort, and re-produced with permission of Auckland Regional

Council.

Veg

eta

tio

n

To Lucas

Creek

General direction of

stormwater flow

Treatment Wetland

StormFilter

Grassy Swales

Raingarden*

Park

ing a

rea

La

nd

scap

e

Foo

tpa

th

Gra

ss

Bu

s s

tation

Sto

rm S

ew

ers

Fen

ce

Ro

ofin

g

*Raingardens, not shown on the map, are located along sections of both Cornerstone Drive and Elliot Rose Avenue


2.3.1 Source Control and use of catchpits

Source control practices are designed to prevent contaminants from entering the

stormwater system. In addition to the use of environmentally conscientious practices

(proper waste disposal and appropriate treatment of pollutant spills, etc.), 71

Enviropod™ filters have been installed within drains and catchpits throughout the car

park, aimed at removing contaminants before they enter the stormwater pipe system.

These filter systems consist of a galvanized steel supporting frame housing a

removable polyester filterbag, which collect litter, organic debris and pollutants

entering the drains, as the stormwater passes through. These filters are an effective,

ARC-recognised, pre-treatment device for use in treatment trains, that allow

contaminants and debris to be removed from the site for off-site treatment and

disposal (for more information, refer to www.enviropod.com)

2.3.2 Raingardens and Swales

Raingardens are located along many of the roads leading into the Albany Busway

including Elliot Rose Avenue and the southern end of Cornerstone Drive. These

raingardens are designed to drain stormwater from the adjacent roads, into

depressions planted with wetland vegetation. Typically, species native to the region

are used (mainly grasses, sedges and Cordyline australis) as they are more tolerant of

the local climatic conditions and are adapted to the prevalent soil and water conditions,

negating the need for fertilizer additions. Raingardens reduce the concentrations of

contaminants entering the watercourse downstream by enhancing absorption to the

soil and encouraging the biological uptake and degradation of contaminants by both

plants and associated microbial communities.

Grassy swales provide a similar service as the raingardens and are located throughout

the car park, spanning a total length of 600 m. These grassed channels (~1.5 m wide)

are also used to separate rows of car parking spaces (without restricting the views

throughout the car park), whilst removing contaminants by natural infiltration,

absorption and enhanced biological uptake.

2.3.3 Filters

One of New Zealand’s largest StormFilters is installed at the Albany Bus Station. The

148-cartridge Stormfilter channels stormwater into an underground chamber and

through a series of rechargeable media-filled cartridges (a mixture of zeolite, perlite and

granular activated carbon) which trap particulates and adsorbs a wide range of

contaminants, including hydrocarbons and heavy metals. These filters are a device

which meets ARC TP10 design for the treatment of total suspended solids and

contaminants associated with heavy vehicular loads. Whilst StormFilters remain

relatively expensive to install, they require infrequent maintenance (every 12-24

months). For more information, refer to www.enviropod.com.


2.3.4 Wetland

A small polishing wetland is located directly downstream of the StormFilter. The

purpose of this wetland is to aid the further removal of stormwater contaminants by

enhancing; (i) the retention, settling and adsorption of contaminants within the

wetland, (ii) the microbial degradation of pollutants, (iii) plant uptake, and also the

degradation of some organic pollutants. Wetlands are of relatively low cost to install

and maintain and add both aesthetic and ecological value to the community green

space.

2.3.5 Lucas Creek

The major aim of the stormwater treatment train located at the Albany Park and Ride

car park is to mitigate the impact of this development on the receiving waters of Lucas

Creek (Figure 2.3). Lucas Creek is a soft-bottomed stream draining a catchment of

approximately 600 hectares across its 16.3 km length. The large catchment area

means that the lower part of the main channel is wider than generally found in North

Shore streams (1-5 m).

Lucas Creek contains abundant koura (Paranephrops planifrons) and populations of

banded kokopu (Galaxia fasciatus), long-finned eel (Anguilla dieffenbachia) and red-

finned bully (Gobiomorphus huttonii). However, the catchment of Lucas Creek is

undergoing rapid urbanization. Future land use will reduce the area of bush to only 2%

of the catchment area, and pasture to less than 1%, compared to a previous cover of

24% in 2005 (NSCC, 2005). Already located within the catchment include Albany

Village, North Harbour Stadium, Albany Mega Centre and Northridge Plaza. New

residential developments are both planned and currently in progress within the Albany

Heights and Fairview Heights areas, and commercial developments continue in the

area surrounding the Albany Mega Centre. As a consequence, there are numerous

areas of recently exposed earth within the catchment. The catchment is bisected by

major roads, including the Northern Motorway (SH1) and Oteha Valley Road, which

runs adjacent to the upper reaches of Lucas Creek.



Location of Lucas Creek and its tributaries, shown in blue (widths of the creek and tributaries are

not to scale). The location of the Albany Bus Station car park is shown in red and the approximate

location of the Lucas Creek sampling site shown in green. Major roads are shown in dark grey.

2.3.6 Bacterial Communities as an Indicator of Water Quality

Biological indicators such as communities of fish and macroinvertebrates have been

widely used to provide an index of overall ecosystem health (Araujo et al., 2000;

Whitfield & Elliott, 2002; Adams et al., 2005; Seilheimer & Chow-Fraser, 2006).

However, a number of recent studies have revealed that the analysis of bacterial

communities (by Automated Ribosomal Intergenic Spacer Analysis - ARISA) can also

provide a sensitive measure of the extent of ecosystem degradation, especially within

highly impacted freshwater streams (Lear et al., 2009a; Lear et al., 2009b). This PCR-

based method creates a fingerprint of microbial community structure from profiles of

the 16S-23S intergenic spacer (IGS) region of the bacterial genome, based upon the

length of the amplified nucleotide sequence, which displays significant heterogeneity

between species. In the present study, community-specific ARISA profiles are used to

provide reliable descriptions of bacterial community diversity and composition (Fig. 2.4)

within the enclosed stormwater channels of the Albany Busway treatment train (where

traditional biological indicators [i.e., fish and macroinvertebrates] cannot be used) and

in the receiving waters of Lucas Creek.

Rather than sampling bacteria within the water column, we assessed communities

associated within microbial biofilms. Biofilms are complex assemblages of

microorganisms within a protective, adhesive matrix of extracellular polymeric

substances, which often account for the large biomass and high diversity of

microorganisms that colonise benthic habitats (Romani & Sabater, 2000). The relatively

sessile nature of microorganisms within the biofilm increases the likelihood that the

abundance of microorganisms within these samples is related to localised influences


within the sample site (with a lower representation from transitory bacteria that are

continually being washed downstream). The composition of these communities is

therefore largely dictated by nutrients, chemical inhibitors, and other growth factors

present in the local environment. This analysis of microbial communities introduces a

broader temporal aspect than can be achieved with simple chemical and physical

monitoring techniques, since the presence of individual organisms are influenced by

past, as well as present, conditions.


ARISA profile of a stream biofilm bacterial community within Lucas Creek. Data are peak length

of the 16S-23S intergenic regions of bacterial genome (x-axis; in nucleotide base pairs) within the

total community, and normalised fluorescent intensity as recorded by a GeneScan automated

DNA fragment analyser (y axis; see appendix 9.2). This method creates a ‘fingerprint’ of the

structure of environmental bacterial communities in which each peak may be considered to

represent a different bacterial taxon, and peak height represents the relative abundance of each

taxon within the total community (note that these assumptions are not strictly true; for a useful

review, refer to Bent et al. (2007)).

0

18

2.3.7 Accumulation of Metals in Microbial Biofilms

Biofilms are known to play a critical role in the transfer of metals and other pollutants

into the foodchain (Rhea et al., 2006; Farag et al., 2007). Indeed, concentrations of

metals are often greater in stream biofilms than sediments (Schorer & Eisele, 1997;

Farag et al., 1998; Holding et al., 2003; Farag et al., 2007) as metals bind strongly to

the reactive surfaces on bacterial cell walls or within exuded microbial polysaccharides.

In addition, metal containing particles (fine sediments, etc.) are trapped within the

microbial biofilm. The strong association between biofilms and metal contaminants

mean that they provide a useful, integrative measurement, of the recent exposure of

the aquatic community to stormwater pollution events.

2.4 Aims and Objectives

We used the site of the Albany Bus Station car park to test the efficacy of a treatment-

train infrastructure in mitigating the environmental impacts of stormwater on the

receiving waters of Lucas Creek. Using the ARISA method of bacterial community

analysis, we address two major research objectives:

Objective 1: Objective 1: Objective 1: Objective 1: Determination of the efficacy of the stormwater treatment train.

1000

1

0

Peak Length (bp.)

Normalised Fluorescent

Intensity

200


We sample biofilm bacterial communities at different locations within the stormwater

pipes of the engineered treatment train. The bacterial community structure within each

sample is then characterised to determine where the greatest differences in bacterial

community structure occur. The aim of this objective is to determine which structures

in the treatment train have the greatest impact in improving the quality of stormwater

discharge into Lucas Creek.

Objective 2: Objective 2: Objective 2: Objective 2: To monitor the impact of the Albany Bus Station car park on the receiving

waters of Lucas Creek.

We analyse bacterial populations above and below the stormwater discharge outlet

from Albany Bus Station into Lucas Creek. Specifically, we compare the similarity

between upstream and downstream communities to determine if significantly different

bacterial populations reside on either side of the stormwater drain, which would

provide evidence of a negative environmental impact caused by the stormwater outlet.


3 Methods

3.1 Site Description

To minimize the impact of the Albany Busway and associated car park on the ecology

of Lucas Creek, a stormwater ‘treatment’ train has been engineered at the site,

including raingardens, grassy swales, a retention wetland and the second largest

stormfilter currently in operation in New Zealand. Our sampling procedure was

designed to monitor changes in biofilm bacterial community structure (a sensitive

biological indicator of ecosystem health) and metal content throughout the treatment

train and also within the receiving waters of Lucas Creek.

3.1.1 Albany Park and Ride Treatment Train

Manholes provided access to underground stormwater pipes throughout the treatment

train (Fig. 3.1). Sites A, B and C drain untreated stormwater from the busway. Site D

drains untreated stormwater originating from Cornerstone Drive. Site E drains

stormwater from the car park and is downstream of sites A, B, C and D. Sites F and G

are located further downstream with additional inputs of untreated stormwater from

Cornerstone Drive. Site H is located at the entrance to the wetland, downstream of the

StormFilter. Finally, site I is located downstream of the wetland. The biofilm bacterial

communities within each sampling site were assessed on 29.01.09 and 26.03.09. The

concentration of biofilm-associated metals were analysed for samples taken on

26.03.09.



Map of the Albany Busway site (36o43’18 S, 174o42’45 E) showing sampling locations (where

manholes provided access to the stormwater drains). Adapted from a map produced by M. Ort

for the Auckland Regional Council.

3.1.2 Lucas Creek

All of the stormwater drained from the Albany Busway is channelled (via the

StormFilter and wetland) into Lucas Creek at the location shown (from a stormwater

pipe originating downstream (left) of the wetland; Fig 3.2). Untreated stormwater

draining Cornerstone Drive and Elliot Rose Road is also channelled into Lucas Creek at

the same location (from the stormwater pipe to the left of Cornerstone Drive; Fig 3.2).

The outlet of this stormwater pipe is at site ‘S’, where the stormwater enters a small

open channel (~5 m in length) before entry into Lucas Creek (Fig. 3.3). Within Lucas

Creek, five sample sites were located upstream of this stormwater outlet (sites 1 to 5)

and five located downstream (sites 6 to 10). Sample sites were located approximately

5 m apart.

Bacterial community structure was obtained for all sample dates and locations within

Lucas Creek. Sediment and biofilm associated concentrations of Cu, Zn and Pb were

analysed for all sampling locations during 30.01.09. Additional samples were analysed

for concentrations of biofilm associated As, Cd, Cr, Cu, Pb, Ni and Zn at sample

locations 1, 10 and the stormwater outlet on each sampling date (30.01.2009, 26.02.09

and 26.03.09). Concentrations of metals within the stream water were assessed on

only 30.01.09.

Grass

Raingarden

Footpath

Bus station

Grassy swales

Storm sewers

B A

D

C

E

H I

G

F

wetland

bypass

To Lucas creek

StormFilter™

N

50 m



Map showing sampling locations (1 to 10) within Lucas Creek (36o43’11 S, 174o42’34 E).

Stormwater from the Albany Busway is channelled into Lucas Creek via a drain running under

Oteha Valley Road (reaching sampling location ‘S’).


Photograph showing sampling site ‘S’ within Lucas Creek, the outlet of a drain channelling

stormwater from the Albany Busway, into Lucas Creek. Sediment immediately surrounding the

outlet is ‘red’ in colour, presumably from deposited metal oxides.

Wetland

10

9

8

7 6

5

4

3 2

1

S

Lucas

Creek

Direction of

stream flow

Sampling locations

Native forest

Grass

Stream

Stormwater pipes

Cornerstone

Drive

(to busway)

Oteha Valley

Road


3.2 Sampling Procedure

Samples were removed from Lucas Creek on three occasions (30.01.09, 26.02.09 and

26.03.09), while samples were removed from the stormwater drain infrastructure of

the Albany Busway on two sampling occasions (29.01.09 and 26.03.09) during this

period. Within Lucas Creek, biofilm biomass was removed from the surface of six

rocks at each sampling site (three for microbiological analysis, and three for the

analysis of biofilm associated metals) using the approach outlined in section 3.2.2. In

addition, one water sample and one sediment sample was also obtained for each

sample site (both c. 50 ml). Within the treatment train, six biofilm samples were

removed from within the stormwater pipes at each sampling site (three for

microbiological analysis and three for the analysis of biofilm associated metals).

3.2.1 Collection of Physico-Chemical Stream Data

Within Lucas Creek, stream physical parameters were recorded during each sampling

occasion to measure spatial and temporal differences in a range of environmentally

relevant parameters. The flow rate of water was measured 2.5 cm above each rock

sampled using a FP101 Flow Probe (Global Water, CA., U.S.A.). Incident light was

measured underwater at the surface of each rock sampled using a photometer (Li-Cor

LI-185B; Design Electronics, Palmerston North, New Zealand). Water temperature, pH

and dissolved oxygen were recorded using a Multi 350i measuring instrument

(Wissenschaftlich-Technische Werkstätten, Germany). Stream depth was also noted.

(The analysis of samples for metal data is described in section 3.3)

3.2.2 Collection and Processing of Biofilm Samples

At each sampling location in Lucas Creek, sample rocks were removed from the water

and biofilm scraped from the entire surface using a fresh Speci-Sponge™ (VWR

International, Arlington Heights, IL, U.S.A.) taking six samples for each sampling

location and date (Fig. 3.4). Samples were similarly removed from the base of

stormwater drains (swabbing an area of approx. 100 cm2 for each sample). Following

biofilm collection, Speci-sponges™ were placed into individual Whirl-Pak® bags (VWR

International, Arlington Heights, IL, U.S.A.) with c. 15 ml sterile water to ensure

complete immersion, and sealed. Samples bags were transported to the laboratory in

darkness, on ice. To separate biofilm biomass from the sponges, samples were

macerated using a stomacher (Lab Stomacher 400, Seward, Norfolk, UK) for 90 s at a

high speed. Sponges were then squeezed to remove the entire sample material and

transferred into centrifuge tubes before the biofilm biomass was pelleted by

centrifugation (8000 g, 20 min)


Figure 3.4. Figure 3.4. Figure 3.4. Figure 3.4.

A Speci-sponge™ is used to remove biofilm biomass. (a) For stream rocks, the entire surface area

of each rock was swabbed using a different sponge. (b) For stormwater pipes, an area at the base

of the stormwater pipe (~ 100 cm2) was swabbed for each sample.

3.3 Processing of Samples for Metal Analysis

To analyse concentrations of metals within biofilm, sediment and water, samples were

sent to Hills Laboratories (Hamilton, New Zealand). Stream water samples were tested

for concentrations of dissolved trace levels of the heavy metals As, Cd, Cr, Cu, Ni, Pb

and Zn, following filtration at 0.45 µm. Samples of biofilm and sediment were dried (at

55 oC) and sieved (mesh size ~250 µM) to remove coarse fractions and to provide a

consistent sample fraction which was then sent for analysis. Biofilm and sediment

samples were analysed for total recoverable concentrations of As, Cd, Cr, Cu, Ni, Pb

and Zn, following nitric/hydrochloric acid digestion using US EPA method 200.2.

3.4 Community Fingerprinting of Bacterial Biomass

DNA was extracted from pelleted biofilm samples within 24 h of collection using a

modified method of Miller et al. (1999). This method combines a bead-beating

methodology with chloroform-isoamyl alcohol extraction, followed by precipitation of

the extracted DNA with isopropanol. Further details of this approach are provided in

Appendix 9.1. The bacterial diversity of biofilm communities, including the unculturable

component, was assessed then using Automated Ribosomal Intergenic Spacer

Analysis (ARISA). This PCR-based method creates ‘fingerprints’ of microbial

communities from profiles of the 16S-23S intergenic spacer (IGS) region of bacteria,

based on the length of the amplified nucleotide sequence. This method enables

sensitive descriptions of community diversity and composition to be attained with a

high level of taxonomic resolution. This method has recently been used for the

evaluation of aquatic bacterial communities (e.g. Jones et al., 2007; Lear et al., 2008;

Lear et al., 2009 a, b, c). Further details of this approach are provided in Appendix 9.2.

a b


3.5 Statistical Analysis

GENEMAPPER software (v 3.7) was used to convert fluorescence data (from ARISA)

into electropherograms, which enable a comparison of the proportional quantities of

different-sized DNA fragments in each sampled community. To visualize multivariate

patterns in community structure based on the bacterial ARISA data, multi-dimensional

scaling (MDS) was done on the Bray-Curtis matrix. All statistical analyses were done

using the PRIMER version 6 computer program (Primer-E Ltd., Plymouth, UK). Further

details of the statistical procedures using in this study are detailed in Appendix 9.3.


4 Results

4.1 Stormwater Treatment Train

4.1.1 Concentrations of Biofilm Associated Metals

Concentrations of zinc exceeded ANZECC interim sediment quality guideline (ISQG)-

High values (ANZECC, 2000) for sediment within every sampling location, reaching a

maximum of 4800 mg kg-1 dry wt. at site E, directly downstream of the car park (Fig.

4.1). Concentrations of zinc declined further downstream to levels similar to sites

located upstream of the car park (1,000-2,000 mg kg-1 dry wt. in sites A, B and C).

Concentrations of arsenic, cadmium, chromium and lead were below ISQG-high values

at all sample locations. However, concentrations of arsenic increased following

passage through the StormFilter (site H and I), reaching a maximum of 29 mg kg-1 dry

wt. (exceeding the ISQG-low trigger value of 20 mg kg-1 dry wt.). Similarly,

concentrations of nickel increased downstream of the StormFilter (sites H and I),

reaching a maximum of 190 mg kg-1 dry wt. at site H, nearly four times greater than

the ISQH-high value for sediment. Concentrations of copper were generally below the

ISQG-high and –low trigger values at every location. Interestingly, there is little

evidence for any decrease in the concentration of metals by the StormFilter treatment

(i.e. the difference between sampling sites G and H) except for zinc, and possibly

copper. Conversely, concentrations of arsenic, cadmium and nickel increased in the

biofilm directly downstream of the StormFilter.

ISQG-values for sediment are provided throughout this study as there are currently no

recommended trigger values for biofilm associated metals. Therefore, because a

biofilm sample exceeds the guideline concentration for sediment may not mean that

the community in the receiving waters are of significant risk of ecological impact.

Nevertheless, it is worth noting that concentrations of zinc observed within these

biofilm samples reach a maximum concentration more than 10 times higher than the

ANZECC ISQG-High trigger value for sediment.


Figure 4.1: Figure 4.1: Figure 4.1: Figure 4.1:

Concentrations of As, Cd, Cr, Cu, Pb, Ni and Zn (mg kg-1 dry wt. of biofilm) at different sampling

locations within the Albany Busway Park and Ride treatment train. Data were collected from

pooled samples, combining the biofilm biomass of three sponge samples, and not replicated. The

dashed lines show the high interim sediment quality guideline (ISQG-High) values (ANZECC

2000) which indicate possible risk to environmental health. The data for sites E to I are plotted as

a line as stormwater passes through each site, sequentially.

0

5

10

15

20

25

30

35

A B C D E F G H I

Ars

en

ic (

mg

kg

-1d

ry w

t.)

Site

ISQG-High = 70

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

A B C D E F G H I

Ca

dm

ium

(m

g k

g-1

dry

wt.

)

Site

ISQG-High = 10

0

10

20

30

40

50

60

70

80

90

100

A B C D E F G H I

Ch

rom

ium

(m

g k

g-1

dry

wt.

)

Site

ISQG-High = 370

0

50

100

150

200

250

300

A B C D E F G H I

Co

pp

er

(mg

kg

-1d

ry w

t.)

Site

ISQG-High = 270

0

20

40

60

80

100

120

140

160

180

A B C D E F G H I

Lea

d (

mg

kg

-1 d

ry w

t.)

Site

ISQG-High = 220

0

20

40

60

80

100

120

140

160

180

200

A B C D E F G H I

Nic

ke

l (m

g k

g-1

dry

wt.

)

Site

ISQG-High = 52

0

1000

2000

3000

4000

5000

6000

A B C D E F G H I

Zin

c (m

g k

g-1

dry

wt.

)

Site

ISQG-High = 410

Arsenic

Lead Nickel

Chromium

Cadmium

Copper

Zinc


4.1.2 Bacterial Community Structure

Multidimensional scaling (MDS) was done to visualise multivariate patterns in bacterial

community structure based on the ARISA data generated from each biofilm sample

(Fig. 4.2 & 4.3). Similar patterns in bacterial community structure were observed on

both sampling occasions where significant differences were detected between

sampling sites (PERMANOVA, P < 0.0001), but not within sampling sites

(PERMANOVA, P = 0.98). For both sampling dates, similar bacterial community

structures were detected for sites A and B, which were significantly different from

sites C and D (PERMANOVA, P < 0.0001). The community within site E, located

downstream of the car park area was significantly different from that of sites A, B, C

and D (PERMANOVA, P < 0.0001). No significant differences were detected in

bacterial community structure between sites E, F and G (downstream of the car park

area), or between sites G and H (which are located at the inlet and outlet of the

StormFilter device; PERMANOVA, P < 0.0001).


Differences in bacterial community ARISA profiles from different sections of the Albany Busway

treatment train sampled in January 2009. Plots are non-metric multidimensional scaling of

bacterial ARISA data, derived from a Bray-Curtis matrix of samples. Letters refer to sampling sites

detailed in Fig. 3.1. 2D stress = 0.16.

Sites ‘upstream’ of car park

Untreated sites ‘downstream’ of car park

After StormFilter treatment

After wetland treatment



Differences in bacterial community ARISA profiles from different sections of the Albany Busway

treatment train sampled in March 2009. Plots are non-metric multidimensional scaling of bacterial

ARISA data, derived from a Bray-Curtis matrix of samples. Letters refer to sampling sites detailed

in Fig. 3.1. 2D stress = 0.20.

4.2 Lucas Creek

4.2.1 Physico-Chemical Stream Data

Concentrations of biofilm-associated metals were greater than concentrations of

sediment-associated metals in every sample obtained (paired t-test, P ≤ 0.05 for Cu Pb

and Zn). Concentrations of Cu, Zn and Pb within the sediment of Lucas Creek were

below ISQG-high and –low trigger values (ANZECC, 2000) at every sampling location

(Fig. 4.4) Concentrations of metals in either biofilm or sediment did not differ

significantly between sampling sites located upstream or downstream of the

stormwater outlet (student t-test, P > 0.05), except for concentrations of biofilm

associated Pb, which were greater in upstream sections (average = 17.8 ± 0.97 mg kg-

1 dry wt., compared to 13.0 ± 0.58 mg kg-1 dry wt. downstream; t-test, p = 0.008). In

addition, concentrations of all metals were generally similar for the stormwater outlet

as for the surrounding sampling sites within Lucas Creek.

Trends in the concentrations of As, Cd, Cr, Cu, Pb, Ni and Zn within Lucas Creek

biofilm samples were broadly similar between sampling dates (Fig. 4.5).

Concentrations of Cd were below detection, whilst concentrations of Ni and Zn were

greater than 50 mg kg-1 (dry wt. biofilm) at all sample locations and exceeded ANZECC

(2000) ISQG-High guideline values. Concentrations of Zn were at least twice as high

within the stormwater outlet than in the sampling locations upstream during January

and March. However, there appeared to be little difference in the concentration of any

biofilm-associated metals between sampling sites located upstream or downstream of

the stormwater outlet (noting that samples were not replicated for any sampling date).

Sites ‘upstream’ of car park

Untreated sites ‘downstream’ of car park

After StormFilter treatment

After wetland treatment



Concentrations of Cu, Pb and Zn (mg kg-1 dry wt.) in sediment (left) and biofilm (right) within: (�)

Lucas Creek sites located upstream of the stormwater outlet; (�) Lucas Creek sites located

downstream of the stormwater outlet; (�) the stormwater outlet. Data for each site were

collected from pooled samples, combining either the biofilm biomass of three rocks or three 50

mL sediment samples and not replicated. Labels on the x-axis refer to sampling location within

Lucas Creek, SWO is stormwater outlet. The dashed lines show the high interim sediment quality

guideline (ISQG-High) values (ANZECC, 2000) which indicate possible risk to environmental

health.

0

10

20

30

40

50

60

70

1 2 3 4 5 SWO 6 7 8 9 10

ISQG-High = 270

0

10

20

30

40

50

60

70

1 2 3 4 5 SWO 6 7 8 9 10

ISQG-High = 270

0

5

10

15

20

25

1 2 3 4 5 SWO 6 7 8 9 10

ISQG-High = 220

0

5

10

15

20

25

1 2 3 4 5 SWO 6 7 8 9 10

ISQG-High = 220

0

200

400

600

800

1000

1 2 3 4 5 SWO 6 7 8 9 10

ISQG-High = 410

0

200

400

600

800

1000

1 2 3 4 5 SWO 6 7 8 9 10

ISQG-High = 410

Biofilm Sediment

Biofilm

Biofilm

Sediment

Sediment

Co

pp

er

(mg

kg

-1)

Le

ad

(m

g k

g-1

) Z

inc (

mg

kg

-1)

Sample Location



Concentrations of As, Cd, Cr, Cu, Pb, Ni and Zn (mg kg-1 dry wt. of biofilm) at different sampling

dates within: (�) Lucas Creek sites located upstream of the stormwater outlet; (�) Lucas Creek

sites located downstream of the stormwater outlet; (�) the stormwater outlet. Data for each site

were collected from pooled samples, combining the biofilm biomass of three rocks and not

replicated. ANZECC (2000) high interim sediment quality guidelines are 70 mg kg-1 (dry wt.)

arsenic; 10 mg kg-1 (dry wt.) cadmium; 370 mg kg-1 (dry wt.) chromium; 270 mg kg-1 (dry wt.)

copper; 220 mg kg-1 (dry wt.) lead; 52 mg kg-1 (dry wt.) nickel; 410 mg kg-1 (dry wt.) zinc.

0

100

200

300

400

500

600

700

As Cd Cr Cu Pb Ni Zn

0

100

200

300

400

500

600

700


0

100

200

300

400

500

600

700


Concentrations of As, Cd, Cr and Pb were all below detection (<0.001 ppm, < 0.00005

ppm, < 0.0005 ppm and < 0.0001 ppm, respectively) in the streamwater and

stormwater outlet (Fig. 4.6). Cu was detected within Lucas Creek, but not within the

stormwater, indicating the Cu is derived from a source further upstream.

Concentrations of both Ni and Zn exceeded 2 mg L-1 water, however none of the

metals exceeded trigger values for the protection of 95% of species in freshwater (1.4,

11 and 8.0 µg L-1 for copper, nickel and zinc, respectively; ANZECC (2000)).

Metal

mg

kg

-1 d

ry w

t.

mg

kg

-1 d

ry w

t.

Metal Metal

January

March

February



Concentrations of As, Cd, Cr, Cu, Pb, Ni and Zn (µg L-1 stream water) (�) within Lucas Creek,

upstream of the stormwater outlet; (�) within Lucas Creek, downstream of the stormwater

outlet; (�) within the stormwater outlet. Data for each site are not replicated.

0

0.5

1

1.5

2

2.5


The average depth of the stream was 20.0 cm and the average pH of the stream water

was pH 7.34 (Figure 4.7; all sampling dates and locations combined). No significant

differences in stream depth, pH or light at the stream bed were detected (student t-

test, P > 0.05) between sampling dates or sections (pooling data obtained from

upstream and downstream sampling locations; sites 1 to 5, and 6 to 10). Significant

differences (student t-test, P < 0.05) in the temperature of the stream water were

detected between sampling dates, being coolest in March (an average of 15.2 oC, all

sampling locations combined), and 0.5 oC cooler within the downstream sampling

locations (6 to 10), all sampling dates combined.

Metal

µg

L-1

str

ea

m w

ate

r



Measurements of physical stream characteristics during (�) January; (�) February; (�) March for

sample sections located (white) within Lucas Creek, upstream of the stormwater outlet; (grey)

within Lucas Creek, downstream of the stormwater outlet; (black) within the stormwater outlet.

0

5

10

15

20

25

30

35

40

0 1 2 3 4 5 6 7 8 9 10 11 12

0

10

20

30

40

50

60

70

80

90

100

0 1 2 3 4 5 6 7 8 9 10 11 12

6.8

7

7.2

7.4

7.6

7.8

8

8.2

8.4

0 2 4 6 8 10 12

10

12

14

16

18

20

0 1 2 3 4 5 6 7 8 9 10 11 12

4.2.2 Bacterial Community Structure


sample sites located upstream, or downstream of the stormwater outlet

(PERMANOVA, P ≤ 0.001; Fig. 4.8). The bacterial community within the stormwater

outlet was marginally more similar to the bacterial communities recorded downstream

of the stormwater outlet, than communities upstream (similarities 43 and 41,

respectively, where PRIMER similarity values ranged from 0 to 100 [perfect similarity]).


stream samples located either upstream, or downstream of the stormwater outlet for

each sampling date (PERMANOVA, P < 0.05), forming distinct groups on MDS plots

(Fig. 4.9). In February and March, there was no difference in the average richness of

bacterial taxa within samples between sampling locations upstream or downstream of

the stormwater outlet (student t-test, P = 0.63 and 0.53, respectively). However, in

January, bacterial taxon richness was significantly higher downstream of the

stormwater outlet (t-test, P = 0.02, for more details of average bacterial taxon richness

between sample sites, refer to appendix 9.6, 9.7 and 9.8)

Depth Light

Temperature

pH

Site

Dep

th (

cm

) p

H

Te

mp

era

ture

(oC

) 1 2 10 9 8 3 4 5 SWO 6 7 1 2 10 9 8 3 4 5 SWO 6 7 L

igh

t a

t S

tre

am

Bed

(L

ux)

1 2 10 9 8 3 4 5 SWO 6 7 1 2 10 9 8 3 4 5 SWO 6 7



Differences in bacterial community ARISA profiles from sections of Lucas Creek, combining data

gathered on each sampling occasion (January, February and March, 2009). Plot is a non-metric

multidimensional scaling of bacterial ARISA data derived from a Bray-Curtis matrix of samples.

Data points refer to samples from (�) upstream of the stormwater outlet; (�) downstream of the

stormwater outlet; (�) the stormwater outlet. 2D stress = 0.21.


Differences in bacterial community data from different sections of Lucas Creek, sampled in

January, February and March, 2009. Plots are non-metric multidimensional scaling of bacterial

ARISA data derived from a Bray-Curtis matrix of samples. Data points are averages (an average of

three rocks per sampling site). Numbers on plot refer to sampling location within the stream. 2D

stress = 0.06, 0.07 and 0.11 for January, February and March respectively.

January February

March Upstream of SWO*

Downstream of SWO*

SWO*

*SWO is stormwater outlet


In Fig. 4.10 bacterial community data collected in this study is compared to data

collected from 18 different Auckland streams, located in both rural and urban

catchments (data from Lear et al. (2009)). Analysis of the ARISA data using multivariate

dispersion values (PRIMER MVDISP) revealed similar variability (dispersion value =

1.020) in bacterial community structure among samples obtained from within the

stormwater treatment trains (in which the greatest physical distance between sample

sites was only 0.3 km), compared to between streams in which the greatest physical

distance was ~100 km between sample sites (Ngakoroa Stream (nr. Pukekohe) and

the Matakana River (nr. Matakana)) (dispersion value = 0.994).

Samples abstracted from sections of Lucas Creek used in this study varied little in

comparison (dispersion value = 0.055) revealing that the stormwater outlet had

relatively little effect on the bacterial biofilm community within Lucas Creek.

Interestingly, samples from lower sections of the stormwater treatment train (e.g., H

and I) were most similar to the communities within Lucas Creek (including sections

located upstream of the stormwater outlet).


Differences in bacterial community ARISA profiles from a range of Auckland streams located

within: (�) predominantly rural catchments; (�) predominantly urban catchments; (��) sections of

Lucas Creek analysed in this study; (�) sections of the Albany park and Ride treatment train;

(Do), sampling sections in Lucas Creek located downstream of the stormwater outlet; (Up),

sampling sections in Lucas Creek located upstream of the stormwater outlet; (SWO), stormwater

outlet from Albany Busway treatment train into Lucas Creek. Letters A to I refer to sampling

locations within stormwater pipes of the Albany Park and Ride treatment train (as detailed in Fig.

3.1). Plot is derived using a Bray Curtis matrix. 2D stress = 0.19.


5 Discussion

5.1 Lucas Creek

This study used the analysis of bacterial biofilm communities to provide a sensitive

measure of the ecological impact of stormwater on a freshwater stream. Our study

revealed statistically significant differences in bacterial community structure between

stream sections located upstream or downstream of a large outlet channelling

stormwater from the site of the Albany Park and Ride car park. However, the observed

differences in bacterial community structure were relatively small, and no differences

in stormwater-associated metals were detected between upstream and downstream

sampling locations. This suggests that the stormwater outlet is currently causing

minimal disturbance to the ecological health of the receiving waters of Lucas Creek.

5.1.1 Accumulation of Metals in the Sediment and Biofilm

ANZECC (2000) threshold values for the protection of aquatic life (in both sediment and

water) were not exceeded for the concentrations of any of the metals monitored in

this study. Although concentrations of biofilm-associated zinc and copper were

elevated within the stormwater outlet, compared to the stream water, the

concentrations of these metals did not increase significantly in the downstream

sections of Lucas Creek, presumably due to dilution by the flow of water from

upstream. Conversely, concentrations of biofilm-associated lead were lower in the

stormwater than in upstream sections of the stream, and were reduced downstream

of the stormwater outlet. This suggests dilution of lead in the receiving waters of

Lucas Creek by the stormwater from the Albany Busway site. Concentrations of

biofilm-associated arsenic, cadmium and chromium were also relatively low within the

stormwater outlet and did not increase downstream of the outlet (compared to

sections monitored upstream).

High concentrations of trace elements have previously been observed to accumulate

within natural biofilm communities exposed to pollutants (Ivorra et al., 1999; Morin et

al., 2008). In the present study, concentrations of biofilm-associated metals were

consistently higher than concentrations of sediment–associated metals. This may have

important implications for aquatic systems, since biofilms are the basis of most aquatic

food webs. Therefore, macroinvertebrate communities, particularly those with scraping

feeding strategies ingest and accumulate biofilm-associated metals (Farag et al., 1998;

Courtney & Clements, 2002), and may provide a concentrated source of metals that

can be toxic to predatory fish and other organisms (Kiffney & Clements, 1993). Since

aquatic organisms at higher trophic levels are directly affected by intimate contact with

microbial biofilm, concentrations of biofilm associated contaminants could provide a

more sensitive measure of the effects of human activity freshwater ecosystem health.

This warrants further study, since at present, there are no recommended guidelines for

acceptable levels of pollutants within microbial biofilms. If, as we expect, the


contaminants associated with biofilm communities are capable of providing a more

sensitive indicator of the impact of stormwater on freshwater ecological communities,

this would provide a significant advance towards the Auckland Regional Councils’ goal

of ‘improving understanding of the cause-effect links between stormwater chemical

contaminants and effects on life in streams, estuaries and harbours’ (ARC, 2009).

5.1.2 Biofilm Bacterial Community Structure

Only minor differences in bacterial community structure were detected between

sampling sites located upstream or downstream of the stormwater outlet. Since no

significant differences were noted in concentrations of stormwater-associated metals,

these differences may have been caused by variation in physical factors including the

reduction in water temperature observed downstream of the outlet, and factors not

monitored as part of this study such as differences in substrate composition and

habitat heterogeneity. The observed differences in bacterial community structure could

also have been caused by certain chemical characteristics of the stormwater, not

recorded in this study (such as concentrations of nitrate, phosphorus, or polycyclic

aromatic hydrocarbons). Finally, the differences in bacterial community structure

between samples sites located either upstream or downstream of the stormwater

outlet could be due to the addition of bacteria in to the stream from the stormwater.

Different communities of bacteria typically inhabit terrestrial and aquatic environments

and during storm events, ‘terrestrial’ communities of bacteria, originating primarily

from sediments surrounding the Albany Park and Ride car park will be washed into the

receiving waters of Lucas Creek. These ‘immigrant’ bacteria will alter the community

composition within the stream. Indeed, in January bacterial taxon richness was

significantly higher in the downstream sample sections, presumably due to the

addition of new populations of the bacteria from the stormwater.

Whatever the cause (urban streams are typically effected by multiple, interacting

stressors (Allan, 2004)), the relatively small differences in biofilm bacterial community

structure and metal content between the upstream and downstream sections of Lucas

Creek provide a strong indication that the stormwater outlet is having little ecological

impact on the receiving waters of Lucas Creek. This suggests that either (i) the

stormwater contains few ecotoxic contaminants, and/or (ii) the ecology of Lucas Creek

is already significantly degraded, such that we observed little effect from the

stormwater outlet.

Our study suggests that the former explanation is most likely since we found no

evidence of harmful concentrations of stormwater-associated metals within the

sediment or water of Lucas Creek. In support of this, Lucas Creek has previously been

identified as having good habitat quality, riparian cover and instream habitat, supporting

a diverse macroinvertebrate community (ARC 2004).

5.1.3 ARISA Methodology

The findings of this study highlight the potential of ARISA to detect differences in

bacterial community structure between complex and varied environmental samples.


The differences in bacterial community structure within sections of Lucas Creek

located upstream or downstream of the stormwater outlet differed very little when

compared to the differences in bacterial community structure between different

Auckland streams. Nevertheless, our ARISA-based technique revealed a remarkable

ability to differentiate between the bacterial communities located either upstream or

downstream of the outlet, revealing them to be significantly different on all sampling

occasions. This supports the use of ARISA as a sensitive and reproducible indicator of

the bacterial community structure within freshwater streams. The sensitivity of this

approach to detect changes in community structure is likely to be improved by the

large number of individual bacterial (many million) analysed within each biofilm sample.

The use of ‘whole-community’ bacterial indicators of stream health offers numerous

benefits compared to traditional assessments of macroinvertebrate and fish

communities, since:

(i) the small sample size required for analysis means that many replicate samples can

be taken from a small sample area (typically the quantity of biofilm obtained from

only 10 cm2 of the stream bed is required for analysis);

(ii) samples are removed from the site with minimal sampling effort;

(iii) samples can be removed with minimal site disturbance, which allows repeated

sampling at the same site, with minimal periods of time required for site recovery;

(iv) samples can be removed from sites in which alternative indicators of stream health

(fish and macroinvertebrates) are not present (such as in the stormwater pipes

examined in this study);

(v) using bacterial ARISA we are able to analyse many hundreds of samples within a

few days, a rate which compares very favourably to high-throughput

macroinvertebrate methods, and

(vi) the approach is cost effective. The costs of sample analysis are $150/sample

(based on the analysis of 16 samples), but are significantly reduced for the analysis

of larger sample numbers ($45/sample, based on the analysis of 96 samples).

Despite the many advantages offered by the analysis of bacterial communities, the

assessment of macroinvertebrate communities remain the favoured indicators of

freshwater ecological health. A key advantage of their use is the ability to use detailed

taxonomic information to provide further estimates of water quality. Indeed, for

bacterial communities, even where detailed taxonomic information has been gathered,

the different functional roles of bacteria within complex environmental communities

remain poorly understood, as does the relative sensitivity of different bacterial taxa to

various anthropogenic disturbances. Additional research is therefore required to

increase the sensitivity of bacterial indicators to gain the maximum potential from the

high-throughput analysis of freshwater bacterial communities. However, as shown in

this study, the analysis of bacterial communities is particularly suited to the

assessment of highly impact environments, in which traditional indictors of stream

health (e.g., fish and invertebrates) are largely absent.


5.2 Stormwater Treatment Train

Although the stormwater outlet did not cause significant differences in the

concentrations of sediment and biofilm-associated metals in Lucas Creek, elevated

concentrations of nickel, and especially zinc were found throughout the stormwater

treatment train, reaching a maximum of 4,800 mg kg-1 of biofilm (dry wt.). Since

concentrations of some metals (most notably of copper, zinc and lead) decreased in

the downstream sampling sections of the treatment train, this would seem to highlight

the effectiveness of the treatment system in reducing the load of these contaminants

entering Lucas Creek.

In support of this, relatively large differences in bacterial community structure were

observed throughout the stormwater treatment train. The bacterial communities within

sample sites A and B were very different from those detected in C and D, which may

reflect differences in contaminant sources between these sites. Sample sites A and B

drain stormwater from the bus shelter and may also drain water originating from SH1,

channelled down the relatively steep decline which leads directly from the highway to

the bus shelter. Sites C and D are located closer to the intersection of Cornerstone

Drive and Elliot Rose Avenue, and contained high concentrations of both lead and zinc.

Interestingly, concentrations of copper, lead and zinc decreased mostly between sites

C to F. Since no ‘in-line’ treatment devices are present between these sampling

locations, reductions in trace metal concentrations are likely to be due to dilution

effects as the piped stormwater is supplemented by additional sources of water

collected from outside of the bus shelter and car park areas. In addition, much of the

stormwater collected at sites downstream of E has passed through some of the 600 m

swale system, before entering the stormwater drain.

In the present study, it is not possible to differentiate the relative effects of

contaminant reduction by different stormwater treatment structures from those

obtained by the dilution (suggested approaches to address this are outlined in section

6.1). Despite the elevated concentrations of especially zinc within the microbial

biofilms of the stormwater treatment train, concentrations of copper, zinc and lead

within the sediment of Lucas Creek were well within interim sediment quality

guidelines (ANZECC, 2000) and concentrations were not significantly different

downstream of the stormwater outlet. This provides evidence that the stormwater

generated from the Albany Park and Ride car park, and surrounding infrastructure, is

not contributing harmful levels of these contaminants to the receiving waters of Lucas

Creek.

The Auckland Regional Council seeks to ‘evaluate innovative processes for removing

chemical contaminants from stormwater’ (ARC, 2009). Interestingly, the StormFilter

treatment device had little effect in reducing concentrations of biofilm-associated

metals throughout the treatment train. Conversely, concentrations of arsenic, nickel

and possibly cadmium increased directly downstream of the StormFilter (between

sample sites G and H), despite there being no additional source of stormwater

between these sampling locations. This would suggest either that: (i) Metals are

leaching out of the StormFilter device (originating from perhaps the filter media, or the

concrete structure of the underground device); (ii) Processes occurring within the


StormFilter are changing the mobility/reactivity of these metals, such that they are

more likely to associate with components of the biofilm further downstream. Indeed,

Fassman et al. (2009) recently reported that concentrations of dissolved solids

increased downstream of the StormFilter. This increase in dissolved solids is of

interest since the behaviour of contaminants in aquatic systems is highly dependent on

the relative distribution of dissolved and particulate forms; the former exhibiting

greater toxic potential as a consequence of enhanced bioavailability. (iii) Additional

sources of these contaminants are entering the wetland. Elevated concentrations of

certain contaminants could be entering the StormFilter via the backward flow of water

from the wetland, which has been reported to occur at this site (Fassman et al., 2009).

Treated wood has been used to construct barriers within the wetland (see appendix

9.10). It is possible that this wood contains chromate copper arsenate [CCA]

preservatives, which could increase concentrations of chromium and arsenic within the

wetland (we note however that concentrations of copper remained low throughout the

wetland system). The origin of the increased concentrations of biofilm-associated

nickel throughout the StormFilter is important to determine since concentrations of

nickel were elevated within the stormwater outlet into Lucas Creek (reaching

concentrations of 190 mg kg-1 biofilm dry wt., four times greater than ANZECC (2000)

ISQG-High values). At present, a large mound of disturbed earth, covering some 2250

m2 is located only 30 m south-east of the wetland (see appendix 9.9). During

significant precipitation events, runoff from this mound is likely to enter the wetland.

However, since the concentrations of contaminants within this soil have not been

monitored, the impact of this mound on the wetland remains unclear.

5.3 General Conclusions

In conclusion, we found little evidence to suggest that stormwater generated from the

Albany Park and Ride car park is having a negative impact on the ecology of Lucas

Creek since: (i) we observed little difference in biofilm bacterial community structure

(used as a biological indicator of stream health) downstream of the major stormwater

outlet, and (ii) we did not observe significant increases in the concentrations

stormwater-associated contaminants downstream of the stormwater outlet.

Concentrations of copper, lead and zinc decreased throughout the stormwater

treatment train. However, no one specific device within the treatment train was

determined to be responsible for this. In addition, the effects of contaminant dilution

via the addition of ‘cleaner’ stormwater at different sections of the treatment train

cannot be quantified.


6 Conclusions and Recommendations

6.1 Implications and Recommendation for Current Management

The concentrations of metals (in both the sediment and water) in the section of Lucas

Creek studied are being maintained within the limits prescribed for healthy freshwater

systems, and did not increase downstream of the outlet. In addition, bacterial

communities, used a biological indicator of stream health, differed little between

samples sites located either upstream or downstream of the stormwater outlet into

Lucas Creek. These findings suggest that current management strategies to minimize

the impact of stormwater originating from the Albany Park and Ride car park, and

surrounding infrastructure on the ecology of Lucas Creek are working well.

However, despite finding little evidence of any negative impact from the car park site,

the ecology of Lucas Creek remains degraded, with reduced fish and invertebrate

fauna in comparison to less impacted reference streams (Parkyn et al., 2005). To

improve the ecological health of Lucas Creek, we recommend additional investigations

along the length of the stream to see if any specific causes of environmental

degradation can be identified, or if the stream ecosystem is instead responding to the

wide variety of diffuse stressors that are common within most urban streams

(including diverse factors such as stormwater contaminants, altered stream hydrology

and connectivity, and losses of riparian vegetation).

To provide a more informative test of the efficacy of treatment trains in mitigating the

impacts of stormwater on the receiving waters, we recommend that more studies

should be undertaken, across a range of study sites, and including locations with

minimal provisions for the treatment of stormwater. These sites will provide a useful

control to determine the relative extent to which the impacts of stormwater are

reduced by engineered structures in the treatment train, as compared to the other

processes (such as additions of less polluted stormwater, which will act to dilute the

relative concentrations of contaminants further downstream in the pipe system). This

would be improved by the appropriate installation of water contaminant and flow

monitoring apparatus, which will help to provide mass balances of contaminant

introduction and loss throughout the stormwater treatment train.

6.2 Recommendations for Future Research

Our study revealed that current management strategies are sufficient to minimize the

impact of stormwater originating from the Albany Park and Ride car park, and

surrounding infrastructure on the ecology of Lucas Creek. However, major

developments are planned for the area directly south and west of the car park site,

which has been zoned for office/residential/retail and entertainment purposes.

Continual monitoring of the site is therefore required to ensure that the current


treatment system is capable of mitigating the harmful impacts of stormwater

originating from these new developments.

No significant differences in bacterial community structure were detected between

sites either side of the StormFilter. In addition, concentrations of biofilm associated

metals decreased little across this treatment device, and the concentrations some

heavy metals actually increased. It would therefore be prudent to further consider the

cost-effectiveness of the installation of these highly engineered structures for the

removal of stormwater pollutants. It would also be desirable to elucidate the

unexpected source of arsenic, chromium, cadmium and nickel in the latter stages of

the stormwater treatment train. Potential contaminant sources that warrant further

investigation include the timber barriers located within the treatment wetland, and the

mound of disturbed soil located to the south east of the wetland. Determination of the

exact contaminant source will aid the better design and maintenance of future

stormwater treatment systems.

Many pollution sensitive macroinvertebrate taxa (e.g. caddisflies, mayflies, stoneflies)

preferentially inhabit (and graze upon) epilithic stream biofilms (Maxted et al., 2003),

which we observed contain far higher concentrations of stormwater-associated metals

than the stream sediment. The concentrations of metals, and other contaminants,

within stream biofilms may therefore provide a better indicator of their effects of

stream ecosystems due to the strong food web links between microbial biofilms and

macroinvertebrate taxa with shredding and scraping feeding strategies. Further studies

are required to determine the reliability of biofilm-associated contaminants as useful

indicator of stream ecological health.

Overall, this study highlights the potential of bacterial ARISA as a rapid and cost-

effective tool to monitor the impact of urban stormwater on aquatic ecosystems. In

support of other recent studies (Lear et al., 2009a, b, c), this study reveals that

bacterial community analysis is a sensitive indicator of ecological health within highly

modified environments in which most traditional biological indicators of water quality

(i.e., fish and macro-organisms) are absent. We therefore recommend the monitoring

of bacterial communities as a part of future studies, which incorporate varied

measures of stream hydraulic, biogeochemical and biotic function to provide an

integrative measure of stream ecosystem health.


7 Acknowledgements We thank the Auckland Regional Council for providing funding and assistance for this

project and the North Shore City Council for their aid in sample collection. Thanks also

to Elizabeth Fassman and Mingyang Liao (University of Auckland) for sharing their

expertise in the operation of the Albany Park and Ride treatment train. Authors Lear,

Ancion, Roberts and Lewis are supported at the University of Auckland by the

Foundation for Research, Science and Technology Grant No. UAOX306.


8 References A.N.Z.E.C.C. 2000. Australian Water Quality Guidelines for Fresh and Marine Waters.

Melbourne, Australian and New Zealand Environment and Conservation Council.

Adams SM, Ryon MG, Smith JG 2005. Recovery in diversity of fish and invertebrate

communities following remediation of a polluted stream: investigating causal

relationships. Hydrobiologia 542: 77-93.

Allan JD 2004. Landscapes and riverscapes: The influence of land use on stream

ecosystems. Annual Review of Ecology, Evolution and Systematics 35: 257-284.

Anderson MJ, Gorley RN, Clarke KR 2008. PERMANOVA+ for PRIMER: Guide to Software

and Statistical Methods. Plymouth, UK, PRIMER-E Ltd.

Araujo FG, Williams WP, Bailey RG 2000. Fish assemblages as indicators of water quality in

the Middle Thames Estuary, England (1980-1989). Estuaries 23(3): 305-317.

ARC 2003. Stormwater treatment devices design guideline manual. Auckland Regional

Council Technical Publication 10.

ARC 2004. Macroinvertebrate community indices for Auckland's soft-bottomed streams.

Auckland Regional Council Technical Publication 303.

ARC 2009. Stormwater: Regional stormwater solutions. Auckland Regional Council.

http://www.arc.govt.nz/environment/water/stormwater/regional-stormwater-

solutions.cfm. Accessed 17 August 2009.

BCG 2009. Auckland regional stormwater project: An action plan to deliver improved

stormwater outcomes, Auckland. Prepared by The Boston Consulting Group,

Infrastructure Auckland and the Auckland Regional Council.

Bent SJ, Pierson JD, Forney LJ 2007. Measuring species richness based on microbial

community fingerprints: the Emperor has no clothes. Applied and Environmental

Microbiology 73(7): 2399-2401.

Courtney LA, Clements WH 2002. Assessing the influence of water and substratum quality

on benthic macroinvertebrate communities in a metal-polluted stream: an

experimental approach. Freshwater Biology 47: 1766-1778.

Farag AM, Woodward DF, Goldstein JN, Brumbaugh W, Meyer JS 1998. Concentrations of

metals associated with mining wastes in sediments, biofilm, benthic

macroinvertebrates, and fish from the Coeur d'Alene river basin, Idaho. Archives of

Environmental Contamination and Toxicology 34: 119-127.

Farag AM, Nimick DA, Kimball BA, Church SE, Harper DD, Brunbaugh WG 2007.

Concentrations of metals in water, sediment, biofilm, benthic macroinvertebrates,

and fish in the Boulder River watershed, Montana, and the role of colloids in metal

uptake. Archives of Environmental Contamination and Toxicology 52(3): 397-409.

Fassman EA, Liao M, Hellberg C, Easton H 2009. Monitoring of the treatment train at the

Albany park n ride. South Pacific Stormwater Conference (29 April -1 May).

Fisher MM, Triplett EW 1999. Automated approach for ribosomal intergenic spacer analysis

of microbial diversity and its applications to freshwater bacterial communities.

Applied and Environmental Microbiology 65(10): 4630-4636.

Holding KL, Gill RA, Carter J 2003. The relationship between epilithic periphyton (biofilm)

bound metals and metals bound to sediments in freshwater systems. Environmental

Geochemistry and Health 25(1): 87-93.


Ivorra N, Hettelaar J, Tubbing GMJ, Kraak MHS, Sabater S, Admiraal W 1999. Translocation

of microbenthic algal assemblages used for in situ analysis of metal pollution in

rivers. Environmental Contamination and Toxicology 2000(37): 19-28.

Jones SE, Shade AL, McMahon KD, Kent AD 2007. Comparison of primer sets for use in

automated ribosomal intergenic spacer analysis of aquatic bacterial communities: an

ecological perspective. Applied and Environmental Microbiology 73(2): 659-662.

Kiffney PM, Clements WH 1993. Bioaccumulation of heavy metals by benthic invertebrates

at the Arkansas River, Colorado. Environmental Toxicology and Chemistry 12: 1507-

1517.

Lear G, Lewis GD 2009. Impact of catchment land use on bacterial communities within

stream biofilms. Ecological Indicators 9: 848-855.

Lear G, Anderson MJ, Smith JG, Boxen K, Lewis G 2008. Spatial and temporal heterogeneity

of the bacterial communities within stream epilithic biofilms. FEMS Microbiology

Ecology 65(3): 463-473.

Lear G, Boothroyd IKG, Turner SJ, Roberts K, Lewis GD 2009a. A comparison of bacteria and

benthic invertebrates as indicators of ecological health in streams. Freshwater

Biology 54: 1532-1543.

Lear G, Niyogi DK, Harding JS, Dong Y, Lewis GD 2009b. Biofilm bacterial community

structure in streams affected by acid mine drainage. Applied and Environmental

Microbiology 75(11): 3455-3460.

Maxted JR, Evans BF, Scarsbrook MR 2003. Development of standard protocols for

macroinvertebrate assessment of soft-bottomed streams in New Zealand. New

Zealand Journal of Marine and Freshwater Research 37: 793-807.

McArdle BH, Anderson MJ 2001. Fitting multivariate models to community data: a comment

on distance-based redundancy analysis. Ecology 82: 290-297.

Miller DN, Bryant JE, Madsen EL, Ghiorse WC 1999. Evaluation and optimization of DNA

extraction and purification procedures for soil and sediment samples. Applied and

Environmental Microbiology 65(11): 4715-4724.

Morin S, Duong TT, Dabrin A, Coynel A, Herlory O, Baudrimont M, Delmas F, Durrieu G,

Schafer J, Winterton P and others 2008. Long-term survey of heavy-metal pollution,

biofilm contamination and diatom community structure in the Riou Mort watershed,

South-West France. Environmental Pollution 151: 532-542.

NSCC 2005. Lucas Creek Stream Assessment; Report No. KC17.

Parkyn SM, Quinn JM, Cox TJ, Broekhuizen N 2005. Pathways of N and C uptake and

transfer in stream food webs: an isotope enrichment experiment. Journal of the

North American Benthological Society 24(4): 955-975.

Ranjard L, Poly F, Lata JC, Mougal C, Thioulouse J, Nazaret S 2001. Characterization of

bacterial and fungal soil communities by automated ribosomal intergenic spacer

analysis fingerprints: Biological and methodological variability. Applied and

Environmental Microbiology 67: 4479-4487.

Rhea DT, Harper DD, Farag AM, Brumbaugh W 2006. Biomonitoring in the Boulder River

watershed, Montana, USA: Metal concentrations in biofilm and macroinvertebrates,

and relations with macroinvertebrate assemblage. Environmental Monitoring and

Assessment 115: 381-393.

Romani AM, Sabater S 2000. Influence of algal biomass on extracellular enzyme activity in

river biofilms. Microbial Ecology 41: 16-24.


Schorer M, Eisele M 1997. Accumulation of inorganic and organic pollutants by biofilms in

the aquatic environment. Water Air and Soil Pollution 99(1): 651-659.

Seilheimer TS, Chow-Fraser P 2006. Development and use of the Wetland Fish Index to

assess the quality of coastal wetlands in the Laurentian Great Lakes. Canadian

Journal of Fisheries and Aquatic Sciences 63(2): 354-366.

Whitfield AK, Elliott A 2002. Fishes as indicators of environmental and ecological changes

within estuaries: a review of progress and some suggestions for the future. Journal

of Fish Biology 61: 229-250.


9 Appendix

9.1 Extraction of DNA from Biofilm Biomass

DNA was extracted from biofilm samples using a modified method of Miller et al.

(1999). Up to 0.25 g of each pelleted biofilm sample were individually resuspended in

270 µl phosphate buffer (100 mM [pH 8.0]), 300 µl SDS lysis buffer (100 mM NaCl; 500

mM Tris [pH 8.0]; 10% sodium dodecyl sulphate) and 300 µl chloroform:isoamyl

alcohol (24:1) within a polypropylene beadbeater vial (containing 0.5 g each of 0.1 mm

and 3.0 mm silica-zirconium beads). Vials were agitated (4 ms-1, 40 s) in a FastPrep

machine (Bio 101, Q-BioGene, Australia), allowed to cool for 1 min and then shaken

once more. Samples were centrifuged (20,000 g, 5 min) and the supernatant (~ 650 µl)

combined with 7 M NH4OAc (360 µl) before being mixed by hand and centrifuged

(20,000 g, 5 min). The supernatant (~ 580 µl) was combined with 0.54 volumes of

isopropanol, mixed, incubated at room temperature for 15 min, and then centrifuged

(20,000 g, 5 min). The DNA pellet was then washed twice with 70% ethanol and air-

dried. The extracted nucleic acids were resuspended in sterile, nuclease-free water

and stored at -80 oC, until analysis.

9.2 Automated ribosomal intergenic spacer analysis of biofilm DNA

The biodiversity of bacterial communities, including unculturable components, was

assessed using automated ribosomal intergenic spacer analysis (ARISA). PCR was

undertaken on extracted DNA using Promaga GoTaq® Green DNA polymerase master

mix (Invitro Technologies Ltd., Auckland, New Zealand) and the universal bacterial

primers SDbact (5’-TGC GGC TGG ATC CCC TCC TT-3’) and LD Bact (5’-CCG GGT TTC

CCC ATT CGG-3’) (Ranjard et al. 2001), with the following amplification conditions: (i)

95 oC for 5 min; (ii) 30 cycles of 95 oC for 30 s, 61.5 oC for 30 s, 72 oC for 90 s and then

(iii) 72 oC for 10 min. To enable analysis by ARISA (Ranjard et al. 2001) the primer

SDBact was labeled at the 5’end with HEX (6-carboxyhexafluorescein) fluorochrome

(Invitrogen Molecular Probes, Auckland, New Zealand). PCR products were purified

(Zymo DNA Clean and Concentrator Kit, Ngaio Diagnostics Ltd., Nelson, New Zealand)

and diluted in sterile water to a concentration of 40 ng µl-1 (using a Nanodrop-8000

spectrophotometer; BioLab Ltd., Auckland, New Zealand). An aliquot of this solution

was combined with 10 µl Hi Di formamide and an internal LIZ1200 size standard

(Applied Biosystems Ltd., Melbourne, Australia), before being heat treated (95 oC, 5

min) and then cooled on ice. To generate ARISA profiles of bacterial community

structure, the samples were then run on a 3130XL Capillary Genetic Analyser (Applied

Biosystems Ltd.) using a 50 cm capillary and standard genemapper protocol [but with

an increased run time (15 kV, 65 000 s)] to record the fluorescent intensity of different

sized PCR products (approximating to the abundance of each bacterial ‘taxon’) within

each sample.


9.3 Quantitative Methods

GENEMAPPER software (v. 3.7) was used to convert fluorescence data (from ARISA)

into electropherograms, which enable a comparison of the proportional quantities of

different-sized DNA fragments in each sampled community. This software was also

used to assign a fragment length (in nucleotide base pairs) to peaks, via comparison

with the standard ladder (LIZ1200; Applied Biosystems Ltd., Melbourne, Australia). To

include the maximum number of peaks while excluding background fluorescence, only

peaks with a fluorescence value of 50 U or greater were subsequently analysed. As

the 16S-23S region is thought to range between c. 140 and 1530 bp (Fisher & Triplett

1999), fragments < 150 bp were excluded from analysis. No samples contained

fragments >1000 bp. The total area under the curve was normalized (to 1.0) to remove

differences in profiles caused by different DNA template quantities, and peak size

rounded to the nearest whole number. Each sample therefore consisted of 850

variables that represent the length (in bp) of the intergenic spacer region of constituent

bacteria, thereby creating a profile of the bacterial community structure within each

sample.

To visualize multivariate patterns in biofilm community structure based on the ARISA

data, multidimensional scaling (MDS) was performed on the Bray-Curtis measure.

MDS is a non-metric procedure that is robust to outliers and preserves the rank orders

of the relative distances among points in the higher dimensional data cloud as well as

possible on a smaller number of dimensions. As well as plotting the relationship

between datasets using MDS, the statistical significant of differences between ARISA

datasets were analysed using permutational multivariate analysis of variance

(PERMANOVA; McArdle et al.(2001)). Statistical analyses were completed using the

Primer 6 (v. 6.1.11) computer program (PRIMER-E Ltd., Plymouth, UK) with the

PERMANOVA+ add-on package (Anderson et al. 2008).


9.4 Bacterial ARISA Profiles – Jan 2009 – Treatment Train Samples

Comparison of ARISA traces obtained from different sections of the Albany Treatment

Train (A to I). Data are peak height (fluorescence; y-axes) and fragment length

(nucleotide base pairs; x-axes). Data are averaged for replicate samples. S refers to the

number of peaks identified (analogous to taxa species richness).

0

18

0

18

0

18

0

18

0

18

0

18

0

18

0

18

18

18

18

18

18

18

18

18

18

0

200 1000

Flu

ore

se

nc

e In

ten

sit

y

Fragment Length (bp)

A

B

C

D

E

F

G

H

I

S = 45

S = 39

S = 65

S = 42

S = 62

S = 56

S = 56

S = 67

S = 63


0

18

9.5 Bacterial ARISA Profiles – March 2009 – Treatment Train Samples

Comparison of ARISA traces obtained from different sections of the Albany Treatment

Train (A to I). Data are peak height (fluorescence; y-axes) and fragment length

(nucleotide base pairs; x-axes). Data are averaged for replicate samples. S refers to the

number of peaks identified (analogous to taxa species richness).

0

18

0

18

0

18

0

18

0

18

0

18

0

18

18

18

18

18

18

18

18

18

A

E

D

C

B

F

G

H

18

I

Flu

ore

sen

ce I

nte

ns

ity

S = 71

S = 78

S = 67

S = 76

S = 83

S = 62

S = 65

S = 70

S = 81


0

18

0

18

9.6 Bacterial ARISA Profiles – Jan 2009 – Lucas Creek Samples

0

18

0

18

0

18

0

18

0

18

0

18

0

18

0

18

18

18

18

18

18

18

18

18

18

18

18

2

1

5

3

4

7

8

9

6

10

Flu

ore

se

nc

e I

nte

nsit

y

S = 63

S = 44

S = 94

S = 64

S = 40

S = 55

S = 69

S = 79

S = 90

S = 97

S = 89


0

18

0

18

0

18

9.7 Bacterial ARISA Profiles – February 2009 – Lucas Creek Samples

0

18

0

18

0

18

0

18

0

18

0

18

0

18

(no data was obtained for sample site 8 in February)

18

18

18

18

18

18

18

0

200 1000

7

9

5

2

4

3

6

1

Flu

ore

sen

ce I

nte

ns

ity

S = 64

S = 72

S = 72

S = 82

S = 46

S = 49

S = 78


0

18

0

18

0

18

9.8 Bacterial ARISA Profiles – March 2009 – Lucas Creek Samples

0

18

0

18

0

18

0

18

0

18

0

18

0

18

18

18

18

18

18

18

18

18

18

18

18

18

0

200 1000 Fragment Length (bp)

7

8

9

5

6

4

1

3

2

Flu

ore

se

nc

e In

ten

sit

y

S = 59

S = 72

S = 40

S = 51

S = 40

S = 41

S = 60

S = 48

S = 43

S = 53


0

18

0

18

0

18

0

18

9.9 Photograph of treated wood barrier at the western, downstream end of the treatment

wetland


9.10 Photograph showing the large mound of disturbed earth south-west of the treatment

wetland

Approximate Location of

Wetland Soil Mound

Date post:	24-Feb-2021
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

Use of Microbial Biofilms to Monitor the Efficacy of a Stormwater … · 2012. 7. 26. · Use of...

Documents